Notice the horizontal axis in the graph below where 20 Hz is on the left and 20,000 Hz on the far right. But the reference scale is optimal and that is why we also measure in this range. Of course, everyone is different and it changes with age, too. The human ear perceives sound frequencies in the range of approximately 20–20,000 Hz. Therefore, in addition to taking into account the total noise level, it is appropriate to analyze which frequencies are more and which are less dominant in the overall sound of a particular cooler or graphics card as a whole (because non-cooling elements, typically coils, contribute to the results as well). This can easily happen, because the new card may have dominant frequencies that do not suit you. Imagine choosing the quietest in a test of thirty graphics cards, because its measured noise level is only “30.1 dBA” (that is, at the limit of what ordinary sound level meters can record), but in the end you find your old “32-decibel” card’s noise a bit more pleasant. And this is a fundamental difference from the user’s point of view. In any case, a standard sound level meter, which has only one value (dBa/dBc) at the output, evaluates it based on a mix of all frequencies, and the result says nothing about whether it is 40 dBA booming or squeaky. But even these usually have a resolution of only 1/3 octave, i.e. That is, unless you currently have an extremely expensive sound level meter with its own spectrograph. But it will not give you a clear idea of how the sound of the measured device sounds. ![]() This quantity indicates whether, under certain conditions of use (measured device/sound level meter), the object of interest is noisy, quiet or something in between. First of all, the disadvantage is that the sound level meter senses the noise level across an entire frequency bandwidth and gives you a value of “sound pressure” at the output. We also use it, but it has several limitations from the principle of its functioning. The easiest way to analyze noise is to use a sound level meter. This typically includes whistling, whirring and similar irritating noises. After all, no one likes computers whose noise level reaches the level of vacuum cleaners, but also those where you, on the one hand, don’t measure high decibels on the sound level meter, but at the same time your ears are burning. It will matter to the vast majority, I dare say. However, this term may also be of interest to those of you who have not yet encountered it (you basically didn’t have a chance to in hardware tests…), but you still care about the sound the computer makes. Longtime readers know well why we began to measure the frequency response of the sound of PC components. ![]() How to read spectrographs and tables correctly Then they won’t tell you much, and that’s a shame if you also care about the exact “noise” value of computer components, or their sound expression. As one of our readers has pointed out – for those who can read spectrographs, they are a gem. In our tests, you will increasingly encounter measurements of the “frequency response” of the sound of PC components. ![]()
0 Comments
|
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |